Thesis (MSc) -- University of Stellenbosch, 2003. / ENGLISH ABSTRACT: A need for intelligent robotic machines is identified. Research and experiments have focussed
on stable, or relatively stable, dynamic simulated systems to demonstrate the feasibility of
embedding advanced AI into dynamic physical systems. This thesis presents an attempt to
scale the techniques to a dynamically highly unstable system - the coordination of movements
in a humanoid model. Environmental simulation, articulated systems and artificial intelligence
methods are identified as three essential layers for a complete and unified approach to embedding
AI into robotic machinery. The history of the physics subsystem for this project is discussed,
leading to the adoption of the Open Dynamics Engine as the physics simulator of choice. An
approach to articulated systems is presented along with the EBNF of a hierarchical articulated
system that was used to describe the model. A revised form of evolution is presented and
justified. An AI model that makes use of this new evolutionary paradigm is introduced. A
variety of AI variants are defined and simulated. The results of these simulations are presented
and analysed. Based on these results recommendations for future work are made. / AFRIKAANSE OPSOMMING: Die beheer van dinamiese masjiene, soos intelligente robotte, is tans beperk tot fisies stabilie
- of relatief stabiele - sisteme. In hierdie tesis word die tegnieke van kunsmatige intelligensie
(KI) toegepas op die kontrole en beheer van 'n dinamies hoogs onstabiele sisteem: 'n Humanoïede
model. Fisiese simulasie, geartikuleerde sisteme en kunmatige intelligensie metodes
word geïdentifiseer as drie noodsaaklike vereistes vir 'n volledige en eenvormige benadering tot
KI beheer in robotte. Die implementasie van 'n fisiese simulator word beskryf, en 'n motivering
vir die gebruik van die sogenaamde "Open Dynamics Engine" as fisiese simulator word gegee.
'n Benadering tot geartikuleerde sisteme word beskryf, tesame met die EBNF van 'n hierargiese
geartikuleerde sisteem wat gebruik is om die model te beskryf. 'n Nuwe interpretasie vir evolusie
word voorgestel, wat die basis vorm van 'n KI model wat in die tesis gebruik word. 'n
Verskeidenheid van KI variasies word gedefineer en gesimuleer, en die resultate word beskryf
en ontleed. Voorstelle vir verdere navorsing word gemaak.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/49756 |
| Date | 12 1900 |
| Creators | Hakl, Henry |
| Contributors | Van Zijl, L., Stellenbosch University. Faculty of Science. Dept. of Mathematical Sciences. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | 97 p. : ill. |
| Rights | Stellenbosch University |
Page generated in 0.003 seconds